How to Fail in Grant Writing

Looking for the fast path to grant rejection?

We provide a list here of proven techniques. We gathered these in the course of serving on grant panels or as program officers, and, in some cases, through firsthand experimentation. We are biologists, but many of our suggestions will be useful to grant writers in all disciplines.

On content:

• Don't explicitly state any goals, objectives, or hypotheses in your grant proposal. A good panelist will be able to figure out your questions from the methods.
• Say that your grant is "transformative"—something the National Science Foundation looks for in particularly outstanding grants; it means that your work will change the approach we take to a particular problem—when it is clearly not. Say that more than once if possible. Heck, go ahead and boldface it! If you claim it is so, it is so.
• However, if your grant is potentially transformative, make it clear in your proposal that you don't know how good an idea you have.
• Make it obvious that you have cut and pasted sections from your other grants into this new proposal. Don't worry if the formatting does not match or there are sentences and sections from the old proposals that have no bearing on this one. Reviewers are impressed by people who are too busy to proofread.
• If your proposal is a resubmission, be snarky about the comments you received from the previous reviewers.
• Use lots of acronyms. Define them several pages after you first use them, if possible, or at least bury the definitions in long paragraphs.
• Don't make any predictions. And if you do make predictions, don't put in any experiments that would actually test them.
• Make sure that the feasibility of your proposal's second and third objectives depends on a particular result from your first objective.
• Don't bother discussing what you will conclude if your data don't turn out exactly as you expect.
• Don't give sample sizes or statistical tests.
• Remember the old axiom: The longer the equation, the better. Panelists will be afraid to acknowledge in front of others that they don't understand it, so they will be more likely to recommend you receive a grant. And remember not to define the parameters of any equations you use. Panelists feel smug when they succeed in figuring it out themselves.
• Be sure to use different symbols for the same parameters in different places in the proposal. Remember to use the same symbols for different parameters in other parts of the proposal.
• When discussing your pilot data in disciplines where "Pn Propose to use a difficult technique (e.g., microarrays, recording from neurons) that you have never done before, but don't offer any assurance that you will have a collaborator. Alternately, propose to use a difficult technique that you have done before, but don't mention your experience or pilot data because, after all, you've done it already.
• Focus your grant entirely on your own study species and/or area of focus. Knowledge for knowledge's sake, right? Dealing with problems of general interest is a waste of time. A good panelist will be able to discern the global impacts of your research without being led by the hand.

On format and style:

• Use weird subheadings that do not map onto one another. For example, begin your proposal by listing Goals 1, 2, and 3, and then label your experiments A through J, with no clear relation to the goals. Reviewers love a challenge.
• Use very few subheadings. Grant reviewers are smart enough to figure out where the subheadings should be. A single multipage paragraph is fine.
• Reviewers love 10-point, Arial-font, single-spaced type. Preferably there should be no spaces between paragraphs, headings, or subheadings, either. Your goal is to leave no white space on the page.
• Use a myriad of type styles. Within a paragraph, try to use BOLD-FACED, ALL-CAPITALIZED TYPE for some sentences, then italicize others, and underline still others. Alternatively, use the same plain style throughout the entire proposal—for headings, subheadings, and paragraphs—for a nice, calming homogeneous appearance.
• Don't use spell-check.
• Don't bother worrying if illustrations or graphs are on different pages than the legends that explain the meaning. Relax, the reviewers can work that out with just a little bit of flipping pages.
• Rely on color alone to distinguish lines from one another in a particular graph. After all, no reviewers will be old-fashioned enough to prefer to read a print copy of your proposal, and then not have a color printer. Program officers don't choose colorblind panelists, either.
• Impress reviewers by using complex illustrations with many panels, arrows, boxes, drawings, and photos. The more stuff you can squeeze in, the smarter you'll look. Condense labels into tiny boxes, so that key parts are unreadable. Also assume that the illustrations are self-explanatory—no need for a pesky extended caption.
• If you are allotted 15 pages for your proposal, use only 12. This is especially effective if you leave out any detail whatsoever about your methods.
• Replace simple, meaningful words with polysyllabic behemoths whenever possible. Don't write "use" when you can say "utilize." Why "use a method" if you can "utilize a methodological technique"? There is no reason to "increase" when you can "exacerbate." Bonus points for using polysyllabic words incorrectly, as in "the elevation in glucose concentration was exasperated during exercise."

On the literature:

• Cite literature willy-nilly. Throw it all in! If possible, give a general statement and then cite a series of people who say conflicting things on the topic. The reviewers will never catch on. They don't care if you understand the literature, just that you know of its existence. It is particularly good if your proposal emphasizes aspects of the literature that are unimportant in justifying your objectives. The reviewers will be impressed that you are so broadly read.
• Alternately, don't cite many papers at all, especially recent ones. The reviewers will assume you know the literature.
• If, in places, your grant says something like "Koala noses are known to be adorable (REF)," be assured the reviewers will understand that you were just too strapped for time to fill in the actual research reference.
• Cite literature that isn't included in the "References" section of your proposal.

On the "impacts" statement:

• If you're applying for an NSF grant, make sure that in your "broader impacts" statement you say that your research on frog metamorphosis will help cure cancer and/or help us understand the function of the human brain.
• Confine your statement about the impacts of your research to things that every scholar would do normally. For example, say you will publish your research and leave it at that.

On your grant-program director and you:

If the grant guidelines ask for names of suggested reviewers, be sure to do the following.

• Suggest only two or three names. After all, the program director should have in mind the very best reviewers for your proposal no matter how obscure your area of research.
• Be sure to suggest names of your closest friends, collaborators, your Ph.D. adviser, or even your spouse. They are the people most familiar with your work, right?
• Never provide university affiliations or e-mail addresses of the names you list. Isn't that what Google is for?

Always keep in close communication with the program director managing your proposal, especially in those critical few days right after the panel meets to review the proposals. Multiple e-mails during that period are OK, but telephone calls really get their attention.

This is also an excellent time to schedule a personal interview with the program director to talk about your grant proposal.

Finally, and perhaps the most important tip of all: Always assume that the panel and the program director will give you the benefit of every doubt.